Solving On-Site Challenges with Point Cloud Data! Promoting DX from Plan Drawing Creation to Maintenance Management

In recent years, the use of three-dimensional point cloud data has attracted significant attention in on-site work such as construction and civil engineering. Point cloud technology can be a powerful solution to issues like labor shortages, operational efficiency, and safety improvement. In fact, the Ministry of Land, Infrastructure, Transport and Tourism is also promoting the use of 3D point clouds through the *i-Construction* initiative, and DX (digital transformation) is advancing across a wide range of tasks, from plan drawing creation to as-built inspection and infrastructure maintenance. This article explains in detail what point cloud data is, how it can solve concrete on-site problems, and how it can be used for plan drawing creation and maintenance management. At the end, we also introduce a new technology anyone can use to start high-accuracy point cloud surveying easily: LRTK.

Contents

• What is point cloud data?

• Background and benefits driving attention to point cloud data use

• Streamlining plan drawing creation with point cloud data

• Using point cloud data during construction

• Using point cloud data for maintenance management

• Challenges and solutions for introducing point cloud data

• Simple point cloud surveying with a smartphone: Introducing LRTK

• FAQ

What is point cloud data?

Point cloud data is a collection of countless points acquired by devices such as laser scanners or photogrammetry. Each point contains X, Y, Z coordinate (position) information, and in some methods attributes such as color (RGB) or reflectance intensity are also attached. By plotting the set of points in three-dimensional space, the shapes of buildings, terrain, equipment, and other objects can be digitally reproduced. Height, depth, and complex structures that are difficult to capture in plan drawings or photographs can be recorded three-dimensionally with high accuracy using point clouds.

Point cloud data is also often referred to as 3D scanning, and it acts like a digital copy of the entire site. Once you scan the site and generate point clouds, you can perform arbitrary dimension measurements and drawing creation on that data. It effectively realizes a state where the site is “always at hand,” and is therefore being used across a wide range of civil engineering and construction processes such as design, construction management, and maintenance.

Methods for obtaining point clouds can be broadly divided into laser measurement (LiDAR) and photogrammetry. Point clouds can be acquired with various measurement devices, including terrestrial and mobile laser scanners, drone-mounted LiDAR, and recently popular SLAM-equipped small surveying instruments or smartphone-embedded LiDAR sensors. Photogrammetry reconstructs 3D shapes from many photos taken with ordinary cameras and converts them to point clouds. The optimal method can be chosen depending on the purpose and site conditions, and the barrier to acquiring point clouds has decreased compared to the past.

Background and benefits driving attention to point cloud data use

The growing industry-wide attention to point cloud data stems from various on-site challenges and expectations that DX can provide solutions. In the construction and infrastructure industries, labor shortages have become serious due to the aging of skilled technicians and a lack of young workers, creating a greater need to operate tasks efficiently with limited staff. Rapid response is also required to ensure safety and quality in the maintenance of aging structures and frequent natural disasters. In this context, the use of 3D point cloud data is expected to be the key to dramatically improving on-site productivity and safety. With government-led DX initiatives providing tailwinds, point cloud technology is no longer a niche advanced technology but is becoming the new on-site norm.

So what specific benefits does using point cloud data offer? The main points are:

• Improved accuracy and coverage: Point clouds from laser measurement or high-precision photo analysis can achieve millimeter-level precision. While manual measurements might capture only a few points, point clouds can cover objects area-wise, making it less likely to overlook subtle deformations or aging effects. As a result, the quality of surveying and inspection greatly improves and rework caused by discrepancies between drawings and reality can be reduced.

• Labor and time savings: Large areas can be scanned in a short time, significantly reducing surveying and measurement work that previously required multiple people over several days. Acquired point cloud data can be automatically analyzed in software, reducing manual quantity calculations and hand-drafting for drawings. This lightens the workload of on-site staff and inspectors, allowing efficient work with limited personnel. It is also effective as a countermeasure to worsening labor shortages.

• Improved safety: Point cloud measurement is basically contactless and remote, so people do not need to enter dangerous areas such as high places, slopes, or busy roads for measurement. Measurements can be taken from a distance even in confined spaces where people cannot enter or in dangerous zones immediately after a disaster. Replacing traditionally hazardous tasks with safe alternatives reduces the risk of workplace accidents.

• Record keeping and data sharing: Acquired point cloud data can be stored as digital records and easily shared among stakeholders using the cloud. Unlike paper drawings or photo albums, it can be retained as objective evidence that does not deteriorate over time. For example, if you archive the point cloud captured at completion, you can quantitatively assess deterioration over time by comparing it with future periodic inspections. Additional dimensions can be obtained later on the point cloud, or cross-sections and elevations can be generated as needed for flexible secondary use. Accumulated 3D data becomes a valuable digital asset for future renovation design or repair planning.

As described above, using point cloud data offers major benefits in quality, efficiency, and safety. In the next sections, we will look at how point cloud technology concretely solves on-site problems in specific tasks.

Streamlining plan drawing creation with point cloud data

One major effect of point cloud data is the streamlining of plan drawing creation. Consider creating as-built plan drawings for buildings or facilities. Traditionally, you would measure interior dimensions one by one with tapes and manual measurement, sketch by hand, and then draft in CAD based on existing drawings. This process requires a lot of time and effort to measure large floors or complex layouts, and it is prone to missed measurements and recording errors. It was also common to overlook wall thickness or column positions and have to revisit the site later.

However, using point clouds from 3D scanning can greatly reduce the effort and mistakes involved in creating plan drawings. For example, if you scan the interior of a building once with a laser scanner or smartphone LiDAR, you will obtain a point cloud for the entire room. Import the point cloud into dedicated software or CAD, create a horizontal slice at a certain height from the floor, and you get a plan projection showing wall and column positions. You then simply trace along that section to produce an accurate plan drawing. Ceiling heights and beam-under heights can also be easily determined by extracting vertical dimensions from the point cloud.

Using point clouds also brings the advantage that “all necessary drawings can be obtained from a single measurement.” Previously, additional measurements were required whenever a section or elevation was needed, but with point cloud data you can cut the model at arbitrary positions to produce cross-sections or generate elevations of the entire building from the side. In other words, once you have a 3D point cloud, you can reproduce all plan, section, and elevation drawings on the data. You can check dimensions from the office without returning to the site, preventing situations like “we forgot to measure that part.” Recreating and utilizing the site in a digital space is precisely what transforms the plan drawing creation process.

Furthermore, creating plan drawings from point clouds contributes to standardizing drawing quality. Tasks that previously relied on individual experience or intuition gain objective backing from scan data, ensuring consistent accuracy regardless of who performs the work. Even non-experts can produce high-precision drawings, reducing dependence on veterans and delivering stable results. As described, point cloud data eliminates waste, inefficiency, and variability in plan drawing creation, enabling fast and accurate drawing production.

Using point cloud data during construction

Point cloud data is powerful not only for creating design drawings but also for on-site management and as-built inspection during construction. For example, before concrete placement for rebar work or for as-built confirmation of structures, supervisors traditionally measured many points with tape measures and levels and compared them with drawings. This conventional method limits measurement locations and carries a risk of overlooking parts of the overall shape. Preparing photos and handwritten records for reporting also took time.

What is being adopted is as-built management using point clouds + BIM/CIM. Scan the in-progress structure in its entirety with point clouds and overlay it on the design BIM/CIM model (3D design data) for comparison. By visualizing the differences between points on the point cloud and the design model, you can immediately confirm whether construction matches the design everywhere. Error distributions can be displayed as heat maps, and pass/fail judgments can be automated, greatly improving inspection efficiency and objectivity. By capturing the as-built condition in terms of surfaces rather than isolated points, point clouds can detect subtle unevenness or deflection that were hard to find previously, raising the level of quality control.

Also, by uploading point clouds to the cloud and sharing them, remote witnessing inspections become possible. Inspections that previously required the client or supervising staff to travel to the site can be confirmed and approved from the office by sharing the acquired 3D data online. For example, upload point cloud data from the site immediately after construction, and the client can view and approve it on a PC at their desk. By discussing the same 3D model via screen sharing in online meetings, near face-to-face confirmation can be achieved remotely. Inspections can be done with zero travel time even for remote sites, and scheduling becomes more flexible. This ability to continue inspections without contact during events like the COVID-19 pandemic or travel restrictions is another advantage, and it represents a true example of new construction management enabled by DX.

Moreover, repeated fixed-point scans during construction assist with progress management and quantity tracking. For example, if you perform weekly drone photogrammetry of an earthwork site and generate point clouds, you can automatically calculate embankment and excavation quantities from the terrain model at that time. Using 3D visuals in progress reports or presentations to clients makes the on-site situation intuitively understandable where 2D drawings fall short. Sharing point clouds allows site representatives and head office technical teams to discuss the same model, improving communication. In this way, using point clouds during construction solves on-site problems in both quality assurance and communication efficiency.

Using point cloud data for maintenance management

Point cloud data is also powerful for infrastructure maintenance and building preservation after completion. For infrastructure such as roads, bridges, and tunnels, monitoring deterioration over time is important. What is being used is periodic 3D scanning monitoring. For example, if you measure point clouds inside a tunnel every year, comparing them with the previous year’s point cloud can capture the progression of fine cracks or changes in cross-sectional shape. Age-related changes that are hard to detect visually by humans can be detected down to displacements of a few centimeters or less (a few inches or less) by comparing point clouds. Early detection of anomalies and planning repairs can help prevent serious accidents.

For bridges and dams as well, if you save point clouds taken at completion or after repair work as a digital archive, you can accurately compare current and past conditions during future inspections. Because point clouds are collections of dimensions, you can make quantitative assessments like “a member has subsided X millimeters since five years ago.” This incorporates objective data into the PDCA cycle of maintenance and pushes inspections beyond paper-based reports toward data-driven maintenance management.

Point cloud use is also progressing in building facility management. For example, large plants and factories have complex piping and equipment, but periodic scans of plant interiors help with renovation work. Constructing a 3D model (digital twin) from the latest point cloud and simulating the fitting of new equipment can detect interferences before construction. Also, if interior equipment inspections are recorded with 360-degree cameras or lasers, details can be checked in the office to prevent inspection omissions and improve work planning. Stacking multiple years of point cloud data enables tracking equipment changes over time and helps update asset management ledgers.

In this way, using point clouds in maintenance management allows you to maintain a state where the real object and the data always match = a digital twin. Sharing inspection results as 3D data smooths communication among stakeholders and enables collaboration where remote specialists can advise based on the data. In future DX-driven infrastructure management, there are plans to integrate point clouds and IoT sensor data in real time and link them to AI analysis for anomaly prediction and automatic diagnosis. As a first step, it is important to digitize visible structural information as point cloud data and accumulate it.

Challenges and solutions for introducing point cloud data

While the benefits of point cloud data are many, some challenges have been pointed out when introducing it on-site. Here are representative challenges and recent trends that serve as solutions.

• Heavy data that is hard to handle: High-density point clouds produce very large file sizes and can slow down ordinary PCs. This is being alleviated by recent advances in PC performance and improvements in dedicated software. Point cloud processing software can decimate unnecessary points or display divided regions, and browser-based visualization on cloud services has emerged. If you extract and save only the portions used for work and thoroughly organize data, the problem of being unable to utilize point clouds due to storage constraints has been largely resolved.

• Cost of equipment acquisition and operation: High-performance 3D laser scanners can cost several million yen or more, making initial investment a barrier. However, you can now choose equipment according to purpose to adjust costs. For example, use surveying instruments for control surveys that require strict accuracy, while using inexpensive 360-degree cameras or smartphone LiDAR for routine site records. Recently, subscription-based point cloud processing platforms using the cloud have emerged, allowing pay-as-you-go use without purchasing expensive software. Companies offering rental equipment or outsourced point cloud measurement services have also increased, making introduction more affordable than before.

• Need for specialized knowledge and skills: Some worry that handling 3D data requires advanced skills. Indeed, specialists such as surveyors or CAD operators were central in the past, but the spread of user-friendly tools has lowered the barrier. Modern point cloud viewers and editing software have intuitive interfaces that users can learn by using them without reading manuals. Automated classification and AI analysis functions also streamline data organization. Training programs and e-learning for on-site personnel are becoming more abundant, so beginners can gradually build skills. The key is not to aim for perfection from the start; try small projects to accumulate know-how, and non-specialists can become proficient.

As such, barriers to point cloud introduction that existed previously are decreasing thanks to technological progress. In particular, the appearance of tools that can easily acquire high-precision point clouds has the potential to solve both cost and skill challenges at once. In the next section, we introduce one representative example: the latest smartphone-based point cloud surveying solution, LRTK.

Simple point cloud surveying with a smartphone: Introducing LRTK

Finally, we introduce the promising new technology LRTK that greatly lowers the barrier to point cloud measurement. LRTK is a positioning system that uses an ultra-compact RTK-GNSS receiver that can be attached to a smartphone. By simply attaching this palm-sized device to your phone, anyone can perform centimeter-class (half-inch accuracy) high-precision surveying on-site. Absolute-coordinate, high-accuracy point cloud measurement that previously required tripod-mounted laser scanners or large drones can be achieved with just a smartphone and a small device when using LRTK.

LRTK features and benefits:

• Immediate high-precision 3D measurement: LRTK acquires point clouds while performing real-time correction of satellite positioning errors using the RTK method, so the measured point clouds are assigned accurate coordinates from the start. There is no need for post-processing to align to control points, enabling immediate on-site comparison with design data and as-built evaluation. Preparation and teardown time for surveying is reduced, allowing quick measurements whenever needed.

• Easy operation usable by anyone: Simply start the dedicated smartphone app and walk around following on-screen prompts—recording survey points and scanning point clouds is completed with one tap. Difficult settings are automated, so even non-specialist surveyors can operate it. Field engineers can take a smartphone + LRTK out of their pocket and quickly scan required areas as needed.

• Compact, lightweight, and highly portable: Since the receiver weighs just over 100 grams and merely attaches to the phone, you don’t need to carry heavy equipment across the site. Surveying is possible anywhere people can walk, including narrow spaces where ladders or surveying staff cannot easily enter. The cable-less, battery-integrated design makes it easy to handle outdoors and minimizes equipment transport and setup effort, realizing genuine labor savings.

• AR features and cloud integration: LRTK is not only for surveying; it also includes design-data-to-field-data AR overlays and layout assistance. You can overlay the 3D design model and the on-site point cloud on the smartphone screen and visualize deviations on the spot. For example, you can confirm rebar positions in AR before concrete placement to prevent construction errors. It also integrates with cloud services, enabling one-touch cloud sharing of acquired point clouds and survey point data. If connected, data can be transmitted to the office in real time from the field, and even outside coverage areas, CLAS-capable models compatible with Japan’s QZSS “Michibiki” can maintain high accuracy, so you can rely on them in mountainous areas.

By adopting LRTK, high-precision surveying and point cloud acquisition that once required specialists’ dedicated equipment become easily accessible to anyone. Even without expensive instruments, being able to scan the site on your own smartphone whenever needed increases the frequency of routine as-built management and equipment inspections, balancing quality control and productivity. It is a device that could become the new norm for point cloud data utilization.

For more details on the LRTK series, please also visit the [LRTK official site](https://www.lrtk.lefixea.com). Case studies, product lineups, and pricing information are available. As a means to start on-site DX from simple measurements, LRTK can be a powerful partner to evolve your company’s workflows to the next stage.

FAQ

Q: What is point cloud data? A: Point cloud data is a collection of many points obtained by laser measurement or photo analysis, each point having position coordinates; it is three-dimensional data. By representing surfaces of buildings or terrain with countless points, spatial information including height and depth can be digitally reproduced. It enables grasping three-dimensional shapes that photos or plan drawings cannot reveal.

Q: How do you create plan drawings from point cloud data? A: Generally, you import the acquired point cloud into dedicated software, cut a horizontal section, and trace its outline to create a plan drawing. For example, if you slice the point cloud at a height of 1 m (3.3 ft) from the floor, you obtain a section (planar projection) that shows the placement of walls and columns. Tracing those lines produces a high-precision plan. Recently, software that automatically converts point clouds into drawings has also appeared, further improving efficiency. With point cloud data, there is no need for additional on-site measurements, reducing oversights and enabling fast, accurate drawing creation.

Q: Don’t you need expensive equipment to introduce point cloud technology? A: It depends on the use case, but you don’t always need to purchase an expensive laser scanner. Recently, affordable devices such as smartphone-embedded LiDAR and small handheld scanners have become common. You can also outsource measurements to surveying companies or rent equipment for the required period. Point cloud processing software is now available as cloud services, allowing you to start without large upfront investments. Solutions like LRTK allow relatively low-cost acquisition of high-precision point clouds, lowering the cost barrier.

Q: Can point cloud data be handled without specialized knowledge? A: Yes. While specialized surveyors or CAD operators were often required in the past, many user-friendly tools are now available. Point cloud viewers enable intuitive rotation and zoom via mouse, and automated analysis functions can classify features or remove unnecessary points with a single click. Although there may be an initial learning curve, many products come with training programs and support. Systems designed for field use like LRTK guide users via clear smartphone app UIs, so basic point cloud measurement and utilization can be started without specialist knowledge.

Q: Point cloud data files are large—how should they be managed? A: High-density point cloud data can indeed reach several GB to tens of GB. In practice, it is common to trim to the necessary area or decimate points to make the size manageable. Modern point cloud software supports cloud storage, enabling large datasets to be shared and viewed over the internet. Downloading only needed parts helps reduce PC load. Many organizations also use NAS or servers for centralized storage and team access. The important practice is to retain the original data while creating reduced-size datasets for practical use. With proper data management rules, storage size does not have to be a major barrier.